A tale of two continents: Infant death clustering in India and sub-Saharan Africa

Description/Abstract

The reasons for death clustering within families and the mechanisms which give rise to the phenomena are of interest to researchers and policymakers. Multilevel discrete-time hazard models are fitted to data from 34 demographic and health surveys from India and Africa to estimate the magnitude of death clustering within families after adjusting for socio-economic and demographic factors. A further stage of the analysis relates death clustering with country-level social, economic, and development indicators. The results show that levels of death clustering are lower in India than in Africa with average intra-family correlation coefficients of 0.01 in India and 0.05 in Africa. A positive association is found of death clustering with poverty, HIV prevalence, overall mortality and fertility, but the association with aggregate income and child under-nutrition is negative. Death clustering is higher in those Indian states with wider gender inequality in literacy.